Live data from Hacker News

Computer Says No

koenvangilst.nl

41–48 of 48 posts

Re: Computer Says No

#41
post #14

The gist ("product becomes a black box") applies to any abstraction. It could apply to high-level languages (including so-called "low-level" languages like C), and few people criticize those. But LLMs are particularly insidious because they're a particularly leaky abstraction. If you ask an LLM to implement something: - First, there's only a chance it will output something that works at all - Then, it may fail on edg…

I agree, there is a reason we settled on programming languages as a interface to instruct the machine. Ultimately it is a tool to indicate our thoughts as precise as possible in a certain domain. People that don't understand the tools they use are doomed to reinvent them. Perhaps the interface will evolve into pseudo code where AI will fill in undefined or boilerplate with best estimates.

> People that don't understand the tools they use are doomed to reinvent them.

Having reinvented them, they will understand. In this way, human progress is unstoppable, even with knowledge micro (and macro) dark ages.

Re: Computer Says No

#42
post #34
post #20

I love the framing here. However, I think what a lot of people don't realize is the reason a lot of executives and business users are excited about AI and don't mind developers getting replaced is because product is already a black box.

I'm more and more convinced top execs are most likely to be advantageously replaced by LLM. They navigate such complex decision spaces, full of compromises, tensions, political knots, that ultimately their important decisions are just made on gut feelings. Replace the CEO with an LLM whose system prompt is carefully crafted and vetted by the board of directors, with some adequate digital twin of the company to projec…

The slow-burning problem is going to be adversarial input and poison data.

Re: Computer Says No

#44

The gist ("product becomes a black box") applies to any abstraction. It could apply to high-level languages (including so-called "low-level" languages like C), and few people criticize those. But LLMs are particularly insidious because they're a particularly leaky abstraction. If you ask an LLM to implement something: - First, there's only a chance it will output something that works at all - Then, it may fail on edg…

This was my take also until recently. It’s now become possible to write better code, faster than you’d get without AI.

It’s all about planning and architecting things before any code is generated. It’s not vibe coding prompt->code, its idea->design->specification->review->specification-> design->architecture->define separate concerns->design interfaces/inter module APIs->define external APIs->specification/review->specification-> now start building the code, module by module. Writing in c++ at Least, I have had very good results with this approach, and I’m ending up with better, more maintainable code, less bugs to hunt, and all in about 0.15-0.25 the time it would have taken to hack something together as an MVP that will have to be rewritten.

Re: Computer Says No

#45
If you do decide to get in touch again, you can ask the agent for assistance understanding the code and you'll get a lot of help. Seems like it's easier than ever?

It's not like working with legacy codebases used to be.

Re: Computer Says No

#46
post #39

Earlier quoted context omitted.

Well - I've learned what a "formant" is today. Looking at the repo, it's not obvious to me what .md is authored by you vs system-generated. This is an observation, not a criticism. I was looking for the prompts you used to specify the papers summarization, which is very nice.

I can do ya one better! https://github.com/ctoth/research-papers-plugin

That is an extremely cool skill

Re: Computer Says No

#47

I really like the framing here (via Richard Sennett / Roland van der Vorst): craft is a relationship with the material. In software, that “material consciousness” is built by touching the system—writing code, feeling the resistance of constraints, refactoring, modeling the domain until it clicks. If we outsource the whole “hands that think” loop to agents, we may ship faster… but we also risk losing the embodied unde…

Hacker News also needs less of your LLM-generated spam
Post reply on HN